Surgery involving the thoracic spine sometimes requires the implantation of metal rods to stabilize and fuse the vertebrae. One increasingly-popular method of anchoring these rods is to place large bone screws through the pedicles that connect the `back'part of each vertebra (i.e. the lamina) with its `front'part (i.e. the body), and then connecting the rods to these implanted screws. Typically a narrow `awl'(or pedicle finder) is forced through the pedicle (creating the pedicle track), and then a screw is placed along this track. Ideally the screw should be fully- contained within bone, but if it `misses'towards the midline (i.e. medially) at any level within the thoracic spine, it may hit and injure the spinal cord. The goal of this study is to develop a new intra-operative monitoring (IOM) test of spinal cord conduction to prevent medial malplacement of thoracic pedicle screws, thereby preventing spinal cord injury. In the first of 2 Specific Aims, we will establish the relationship between the amount of energy needed to electrically stimulate spinal cord and nerve roots through the pedicle track (i.e. prior to screw implantation) and the position of that screw relative to the pedicle, once the screw is placed. We anticipate that pedicle tracks which have broken through the bone of the medial pedicle wall will lead to spinal cord stimulation with weak current intensities. We will retrospectively develop rules to tell us when it is safe to put a screw in a particular pedicle track. In Specific Aim #2, we will apply these rules in a prospective manner to actively prevent screw malplacement during surgery in new subjects. Moreover, we will randomize these subjects into two cohorts. In one group we will actively stimulate through a modified pedicle finder while the surgeon is making a pedicle track, to provide immediate feedback to the surgeon if the orientation of the pedicle track needs to change. We anticipate that there will be fewer cases of medially-malplaced thoracic pedicle screws in subjects studied in Specific Aim #2 compared to those tested with the protocol from Specific Aim #1. Moreover we expect that those subjects who receive active stimulation through the pedicle finder will have the lowest incidence of medial screw malplacement of all subjects tested. By the end of this study period, we will have developed and validated a novel form of intra- operative monitoring for thoracic pedicle screw placement. Not only will this test lower the risk of spinal cord injury, but it will also lower the numbers of subjects who require additional surgery to revise screw placements that " while not causing spinal cord injury per se" are still encroaching upon the canal space enough that they would eventually lead to symptoms if left in place. PUBLIC HEALTH RELEVANCE: Surgery to the thoracic spine to treat deformity or trauma places the spinal cord at relatively high risk for injury, particularly when screws are implanted into the pedicles of the spine. This project will develop a novel method to prevent accidental screw placement into the spinal cord, thereby preventing spinal cord injury.